Gianluigi Lopardo's repositories
attention_meets_xai
Code for the paper "Attention Meets Post-hoc Interpretability: A Mathematical Perspective"
anchors_text_theory
A Sea of Words: An In-Depth Analysis of Anchors for Text Data
Open-World-Recognition
The project's goal is to get familiar with cutting-edge models capable of acting in an open world, incremental learning approaches in image classification and open set strategies
anchors_vs_lime_text
A Comparison of Feature Importance and Rule Extraction for Interpretability on Text Data
HELOC-Credit-Approval
This notebook is ispired by the AIX360 HELOC Credit Approval Tutorial, which shows different explainability methods for a credit approval process. Here XGBoost is used for classification, achieving better accuracy than most of the models used in that notebook. Then, feature importance methods are shown, to be compared with the Data Scientist explanations methods provided in the above notebook. The first ones come directly with XGBoost and the other is based on SHAP.
Reinforcement-Learning
My solution of a simple Reinforcement learning problem
Absenteeism_prediction
Prediction of absenteeism at work, with machine learning models for classification.
CBC-ratings_prediction
The purpose of this paper is to analyze how and how much a film's attributes affect its rating, using several regression techniques.
Financial-articles_analysis
(Part of) a project for the Business intelligence class. The goal is to analyze a dataset of financial articles and apply machine learning technique to extract useful information.
pcs_project
Scientific programming and computing project carried out in MATLAB.
Pilgrim-dropout_prediction
The purpose of this work is to predict which customers are about to leave the bank. To do this, the main classification algorithms will be used to predict whether a customer from 1999 will still be a customer in 2000 or not.
Statistical-Models
A series of statistical models applied to different case studies
Transfer-Learning
The aim of this project is to apply and explore Transfer Learning. The dataset used is Caltech101, the neural network used for the first part of the project is AlexNet.
German-Credit-Data_credit_risk
This project is about the analysis of credit risks of German Credit Data. Different classification models and preprocessing methods are used and compared.